The RAG King 👑 goes HUGE!!!
Вставка
- Опубліковано 7 сер 2024
- 🔗 Links 🔗
txt.cohere.com/command-r-plus...
Today, we’re introducing Command R+, our most powerful, scalable large language model (LLM) purpose-built to excel at real-world enterprise use cases. Command R+ joins our R-series of LLMs focused on balancing high efficiency with strong accuracy, enabling businesses to move beyond proof-of-concept, and into production with AI.
Command R+, like our recently launched Command R model, features a 128k-token context window and is designed to offer best-in-class:
Advanced Retrieval Augmented Generation (RAG) with citation to reduce hallucinations
Multilingual coverage in 10 key languages to support global business operations
Tool Use to automate sophisticated business processes
- Cohere Announcement
❤️ If you want to support the channel ❤️
Support here:
Patreon - / 1littlecoder
Ko-Fi - ko-fi.com/1littlecoder
🧭 Follow me on 🧭
Twitter - / 1littlecoder
Linkedin - / amrrs - Наука та технологія
closed source is temporary, open source is eternity
I hope that was the case - but time will tell
@pixelperfectpravin in long battle open source Ai will eventually beat closed Ai
We need a RAG Tutorial for this!!!
I used it just now. was amazing!!!
Can you do a dive into DSPy versus SGLang? Seems that DSPy is getting all of the attention but that the SGLang approach might be simpler but just as effective. Cheers!
How does it compare with GPT3.5-turbo? The king of pricing is 3.5-turbo
I tried it and its way better than GPT-4 or Claude opus. Lets go open-source 🔥
Any particular test or prompt that stood out ?
Way better in what sense? It's RAG capabilities or it's general intelligence?
"better" for what?
Anything other than commercial purpose? ( Well, thanks for the video)
I gave it a roughly 6,000 token prompt and got a response of a few thousand tokens, it made 2 mistakes. Close but no cigar
Did you try the same with Gemini 1.5 pro?
No I should though@@1littlecoder
Yeah Gemini 1.5 Pro has a very impressive recall (from what I've seen) so would love to see how it goes through this!
Yeah I didn't do full audit but seems to have nailed it, its answer was shorter and less verbose as well which may have helped@@1littlecoder
It made one mistake, sigh, so close
EDIT: 2 mistakes
Nice model. Not so good license.
Goes huge? What… does that mean and when in the history of the English language has that ever been used